The goal of this project is to develop a very low-cost, whole-head neonatal MEG system that can adapt to the shape and size of the head. Clinically, there is a longstanding need for a technology that can aid in the diagnosis and prognosis of babies with brain dysfunction due to intrauterine and postnatal central nervous system insults, which may result in mental retardation and neurodevelopmental disorders. The proposed system is completely non-invasive and is based on a recent breakthrough in atomic magnetometer technology that eliminates the majority of the costs and complexity associated with the use of superconducting sensors in traditional MEG systems. In the Phase I project, we identified and addressed all major technical risk elements, and successfully demonstrated the viability of our approach. In Phase II we will develop a whole-head MEG system for neonates and will demonstrate accurate source localization.